CSV Merger

How to combine CSV files

1. Add two or more CSV, TSV or TXT files, or paste delimited text. Check each source’s separator and column names.
2. Keep strict matching for the same columns, or explicitly include all columns and review where empty cells will be inserted.
3. Preview the combined rows, then download the full CSV and a compact JSON merge report.

What the merger does

• Aligns exact column names across files. A source with name,id maps into the first source’s id,name order.
• Optional all-columns mode keeps first-seen names and inserts empty cells only where a source lacks a column.
• Keeps source and row order, leading zeros, whitespace, multiline cells and duplicate rows.
• Accepts separate comma, semicolon, tab or pipe input separators and exports with your chosen separator, line ending and optional UTF-8 BOM.

When to use it

• Stack monthly exports whose columns are reordered.
• Combine contact lists with some columns present in only one file.
• Collect several local reports before checking duplicates in a separate tool.

Understand the result

Strict mode requires the same exact column names in every source. Email and email are different. The first source sets output order; id,name plus name,id produces one id,name table.

All-columns mode adds new names in source order. If two rows lack price and one row lacks name, it inserts three empty cells. The JSON report identifies absent columns; plain CSV cannot distinguish them from original empty values.

Without headers, all sources must have equal widths and you must confirm that positions have the same meaning. No artificial header is exported. Every input record becomes one output row.

This app appends rows; it does not join by ID, rename columns, sort or remove duplicates. A sep= directive is treated as data. CSV quoting does not prevent spreadsheet formulas, and spreadsheets may change long IDs or leading zeros on import. The report has mappings and row ranges, not raw values or a backup of original formatting.

Privacy and resource limits

File contents are processed locally in your browser. You can use 2–20 sources, each file up to 10 MiB and retained files up to 20 MiB total. All sources together are limited to 5 million text units, 100,000 records, 2 million input cells and 256 columns; output is limited to 2 million cells and each download to 25 MiB. The JSON report may contain sensitive column names.